PSN-L Email List Message

Subject: Re: Relative Mag Calc question
From: John or Jan Lahr JohnJan@........
Date: Mon, 26 Mar 2007 08:29:32 -0700

Sounds like you're doing well.  Don't be surprised if there is quite 
a variation between your computed magnitude and the USGS value.  For 
large quakes they will be quoting moment magnitude (Mw), not 
mb.  There are always variations between the various scales (Mw, mb, 
Ms, etc).  Also for any given mb or Ms magnitude reported, the value 
given for the event will be an average over all of the stations 
reporting.  The station to station values can vary by as much as 0.5 
units; some of this is due to the "radiation pattern," which causes 
some directions to receive more energy than others.  To "back 
calculate" to find your station sensitivity, it would be best to do 
this for a large number of earthquakes and then average the 
numbers.  Even if your calibration is correct, you can expect to find 
variations of a few tenths of a unit from the reported magnitude .

To find the USGS estimates for the various magnitudes that have been 
computed for recent events, go to:

INTERIOR                                                            NO. 7-085
MAR 26, 2007

UTC TIME    LAT     LONG    DEP GS MAGS Q SD  GAP STA           F-E 
HRMNSEC                         MB  Msz          USED

MAR 25
004002.1  20.666S 169.426E  35G         A 0.9  39 192 VANUATU. MW 7.2 (GS).
004157.3  37.281N 136.602E   5G         A 0.9  33 115 NEAR THE WEST 
For some reason, mb and Ms are not given for these recent large 
events.  Perhaps this is a new policy.

At 12:46 AM 3/26/2007, you wrote:
>Hi John,  thanks,  the counts per micro meter was my quest??    so, 
>I plugged in the guess-ta-calculated 29 for my system response, 
>along with the other data and came up with 7.1Mb for the first 
>Vanuatu event,  and 6.9Mb for the second Vanuatu event!    I'll be 
>looking to see how I compare to a USGS update for the second 
>event,,  for now they are calling it a 6.0????   My system has about 
>a 4.5 second natural period = .2222 hz, so the surface wave 
>component was only about 10 count, which I subtracted from the peak 
>count for the calculation!    For the Japan quake, I guess-ta-mated 
>and subtracted the Vanuatu component and came up with 
>6.8Mb,,,   (famous last words,,  I think I understand a little 
>better now,, ha!!)
>Also, it looks like my first problem concept  (log of count 
>ratio)  is a reasonable quick estimate to compare events with the 
>same location and depth????
>  Thanks again,
>  Stephen


Public Seismic Network Mailing List (PSN-L)

[ Top ] [ Back ] [ Home Page ]